Communication and Energy Efficient Slimmable Federated Learning via Superposition Coding and Successive Decoding
This addresses energy and communication efficiency for mobile devices in federated learning, offering an incremental improvement over existing methods.
The paper tackles the problem of heterogeneous energy capacity and communication throughput in federated learning on mobile devices by proposing SlimFL, a framework using slimmable neural networks and superposition coding. Simulation results show SlimFL trains 0.5x and 1.0x models simultaneously with reasonable accuracy and convergence, using 2x fewer communication resources than vanilla FL, and achieves higher accuracy with lower energy in poor channel and non-IID conditions.
Mobile devices are indispensable sources of big data. Federated learning (FL) has a great potential in exploiting these private data by exchanging locally trained models instead of their raw data. However, mobile devices are often energy limited and wirelessly connected, and FL cannot cope flexibly with their heterogeneous and time-varying energy capacity and communication throughput, limiting the adoption. Motivated by these issues, we propose a novel energy and communication efficient FL framework, coined SlimFL. To resolve the heterogeneous energy capacity problem, each device in SlimFL runs a width-adjustable slimmable neural network (SNN). To address the heterogeneous communication throughput problem, each full-width (1.0x) SNN model and its half-width ($0.5$x) model are superposition-coded before transmission, and successively decoded after reception as the 0.5x or $1.0$x model depending on the channel quality. Simulation results show that SlimFL can simultaneously train both $0.5$x and $1.0$x models with reasonable accuracy and convergence speed, compared to its vanilla FL counterpart separately training the two models using $2$x more communication resources. Surprisingly, SlimFL achieves even higher accuracy with lower energy footprints than vanilla FL for poor channels and non-IID data distributions, under which vanilla FL converges slowly.